• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

how to eq harsh upper mid - high frequencies?

thanks for this vid! I'm more confident now with these speakers :) I never tried going mid all up on the knobs, gonna try that tonight!


I've got a wooden ceiling with 'open' beams that break the otherwise flat ceiling. Is this good or bad acoustically? Should I maybe try playing absorbers between the beams?
I would think the corner/left wall is your issue. You could absolutely do something about the ceiling, but... you know, the first wall reflection from the left speaker is going to be very close to the speaker itself, so if you could put something that is at least sonewhat effective down to ~500hz behind the speakers and extending to the door on the left... I'm no acoustics expert, but I really think that is what you should focus on, and PEQ the bass.

Edit: I should add that I currently have a similarly asymmetrical speaker setup, left speaker in a corner just like that. I have measured, listened and EQ'd A LOT and it never get quite right.
 
I was having a similar problem recently which was mainly attributed to the Tannoy speakers I had, vocal vowel sounds such as ‘hey, you’ or ‘me’ along with saxophones and some guitars were like ice picks to the ears at times. Since swapping to some much more natural sounding KEFs I don’t have that Issue anymore, but I do still occasionally have the want to tame the presence region slightly on a select few recordings now and then.

Using a Schiit Loki Mini+ reducing the 2KHz dial by 1.5 - 2db is all I’ve found that is needed to take away some of the edge of some recordings without dulling the overall sound. Using the Schiit is like a bit like using a broad brush stroke as the centre bands have an effect at quite some way at either side of the centre frequcies but subtle adjustments seem to work nicely without changing the overall character of the recordings. I find some troublesome ones become more palatable during extended listing sessions.

I don’t think your room is at fault, it doesn’t look terrible to me. It’s either the speakers (maybe/maybe not since you’ve tried others) or it’s your ears. Whatever means of EQ you have I would try it using subtle (no more than 2db) adjustments from 2KHz - 4KHz and live with it for a while to see how you get on. Changes in 3db or more are going to sound drastic IMO, try being subtle and give yourself some time to adjust. You might find after an hour or so that switching back to flat seems too intense.

If 2KHz - 4KHz isn’t working try going up to 6KHz. I think any higher will start to sound rolled off and most harshness tends to be the presence region from my own experiences over the years.

How is it with the speaker grilles in place?
 
You’re going about it the wrong way! ;)

Get yourself one of these and you’ll never have to worry about harsh highs again.
You can bathe in liquid rich sound:

1736393300670.jpeg


(don’t worry I’ll see myself out the door….)
 
Measurements?

Helps with the guessing.
 
Just a thought, get some 2-4" thick pieces of light to medium density soft foam and put it on either side of the mid and high drivers and one on top of the tweeter (the closer the more directivity)

They need to be fairly close. The closer the more the driver will lose its dispersion. The GAIN in the room from untreated walls and directing the waves might just do the trick. MANY older speakers had foam added to the outside edges of the drivers vs recessing the driver into a waveguide. I'm sure it will help and there used to be a reason for L-Pads.

Few if any used to treat rooms. L-Pads, foam and even a very light see through foam or cheesecloth in front added to a HOT tweeter or mid driver can help.

You can add that vail to the speaker grill if you're worried about looks. BUT you can experiment to get close with different thicknesses and how tight the weave is on
different material.

To be quite honest, it used to be a huge problem before. Especially with the throat of a tweeter horn setup. They would line the throat with shag carpet it was so bad in some cases with ZERO room treatment. There is a reason for tone controls and L-Pads. No two rooms are the same, not to mention young or sensitive ears. BTW if you get used to it that means you're going deaf.

One more thing, HOT MIDS can really add to that boil. It's one of the main reasons I use EL34s for the mids in most of my setups. I can deal with most tweeters as my
ears have aged and mellowed. The combination of small planars/ribbons with the right valve is easier on the ears but still has crystal clarity and NO BOIL.

It reminds me of the days of metal domes and all the hot dog tweeters like Wilson and B&W used. I hated the GD things. It took a VERY well-treated room to make them work for my ears until I was well into my LATE 50s.

The test is simple, can you sit for more than 30 minutes without getting uncomfortable? THAT is the measure you're looking for.

I seldom have to use any EQ IF you understand how and what a room is doing. Untreated walls are a crap shoot no matter what you use for room correction.
I've NEVER needed to use DSP or even EQ, once I understood the way drivers and how they are mounted actually works in a given room.

Best of luck.

Regards
 
Are you using Roon for music?
If so, its DSP engine can do everything including convolution
 
I would suggest that before any investigation of potential cures one has to find the cause of the perceived harshness first. So let us go through the chain:
1 Are you sure the harshness is not part of the music itself? I ask because whenever I attend concerts of large symphonies, choral works or opera, even in superb acoustics like the Amsterdam Concertgebouw, there can be shades of harshness. Real music is not invariably sweet and mellow. So how does it compare with the sound of your local symphony orchestra?
2 Well designed electronics do not normally have a sonic signature, but there is one exception, and it may be relevant here, precisely because it causes harshness: clipping. This can happen in the input stage when the output of e.g. a CD player exceeds the input sensitivity of the amplifier. This was a major problem in the early days of the CD, but it can still happen when the gains are not properly matched. An inline attenuator is the simple solution. The other form of clipping is from insufficient amplifier power. This may clip the signal on dynamic peaks. When I moved from 2x45 watt in a large room to 2x140 watt the harshness in louder passages was almost completely gone. An RME AD-2 DAC replacing my old Quad 33 preamp did the rest - it had so much more headroom to cope with digital sources.
3 Room acoustics. An insufficiently damped room will generate a lot of reflections at higher frequencies. Acoustic treatment really is the only good solution (rugs, curtains and book cases may suffice). I am a great fan of dsp room eq, but only for the frequencies below the Schroeder frequency or a bit above, say 300 Hz, depending on room size. Of course, one could equalize the peaks above that, but the snag is that the listening position that benefits from the room response equalization gets narrower and narrower, the higher the frequency. Move your head a few inches or less, and you are equalizing the wrong frequencies.

Of course, proper measurement should be part of the investigation. and for all we know, these old speakers may have a seriously flawed frequency response.
 
I now see that you have already answered a lot in your earlier thread from a year ago. From what I read there your problem is the room, and absorption was the answer. So find some nice looking panels and you are probably done.
 
thanks for this vid! I'm more confident now with these speakers :) I never tried going mid all up on the knobs, gonna try that tonight!


I've got a wooden ceiling with 'open' beams that break the otherwise flat ceiling. Is this good or bad acoustically? Should I maybe try playing absorbers between the beams?

First that I would try is extreme toe in. Like, having the left speaker pointed to your right ear and vice versa. The goal here would be in having the left side room boundaries less illuminated by the left speaker. This way you may also benefit in having a wider sweet spot, so less perceptual timbre changes when you move your head or not sitting dead centered at your listening position.

If you care to do so, you may have to position the speakers a bit closer together. It's all about manipulating the early reflections so that the measured magnitude (and phase) response would show less difference in between the channels. The more the loudspeakers are spaced and aimed right, the more coherent center phantom image you would get, also a wider sweet spot, so a larger area in which you get good imaging. The most obvious quality would be improved transient response and more clarity in the mid-bass. This is a slow process and takes time in measuring each speaker individually and then both in order to extract meaningful information in what to do next. Remember that the loudspeakers have a job of working in unison for having a good stereo reproduction.

After aiming and spacing is right, confirmed by measurements and listening tests, then it would be more likely that DSP would also yield more consistent results. Basically, it would be more clear what to do next.

As far as acoustic treatment, some treatment on the floor (even as little as having some soft mats in front of speakers) also some absorption on the ceiling may help. Having a door to your left, I would suggest extreme toe in I described above and listen to the timbre changes by having someone else opening and closing the door. The less energy you send sideways, the less the early lateral reflections would be in level.
 
So I got the miniDSP Flex and did a lot of testing. First I tried the REW EQ and then the Dirac. The results are really nice, the bass is so much better now, but the highs are still harsh. I have to use -db in PEQ after REW EQ or Dirac to tame them. But after that the sound gets "muffled". I also somehow get better results with the REW EQ. Maybe because I can adjust the EQ much faster than with Dirac (it's a pain in the ass to connect/disconnect the Dirac program all the time when I just want to adjust the EQ curve).

you know, the first wall reflection from the left speaker is going to be very close to the speaker itself, so if you could put something that is at least sonewhat effective down to ~500hz behind the speakers and extending to the door on the left... I'm no acoustics expert, but I really think that is what you should focus on, and PEQ the bass.
i had 2 GIK basstraps behind the speakers in the corner for testing and REQ measurements made no big difference (at least to me, i'm not good with REW). Here is the measurements with and without the basstraps if anyone cares: GIK measurements

I would suggest that before any investigation of potential cures one has to find the cause of the perceived harshness first.
I see what you mean but it really is unlistenable without hard -db EQ in the 1-5 kHz region.

From what I read there your problem is the room, and absorption was the answer.
Yes it's definitely my room but i did the absorbing wall now (which helped a bit) and that's all i can do unfortunately in the living room.

tomorrow I will set up my desktop computer in the living room and try Dirac again. It's very tiring on the small macbook display. If anyone got some tips for me for testing/measuring or Dirac please enlighten me, i'm so lost with all of this (-:

PS: I moved the couch a few days ago and put the speakers in a symmetrical arrangement to see if it would help, but nope, same problem. Here is a photo of it:

20250109_150746.jpg
 
If you're satisfied with the bass, try creating four slightly different curves for the midrange and higher frequencies within Dirac, then save them to four different slots on the miniDSP. Switch between them until you find one you prefer. Once you identify your favorite, use it as a template for further adjustments until it sounds the way you want with tracks you're familiar with.
 
So I got the miniDSP Flex and did a lot of testing. First I tried the REW EQ and then the Dirac. The results are really nice, the bass is so much better now, but the highs are still harsh. I have to use -db in PEQ after REW EQ or Dirac to tame them. But after that the sound gets "muffled". I also somehow get better results with the REW EQ. Maybe because I can adjust the EQ much faster than with Dirac (it's a pain in the ass to connect/disconnect the Dirac program all the time when I just want to adjust the EQ curve).


i had 2 GIK basstraps behind the speakers in the corner for testing and REQ measurements made no big difference (at least to me, i'm not good with REW). Here is the measurements with and without the basstraps if anyone cares: GIK measurements


I see what you mean but it really is unlistenable without hard -db EQ in the 1-5 kHz region.


Yes it's definitely my room but i did the absorbing wall now (which helped a bit) and that's all i can do unfortunately in the living room.

tomorrow I will set up my desktop computer in the living room and try Dirac again. It's very tiring on the small macbook display. If anyone got some tips for me for testing/measuring or Dirac please enlighten me, i'm so lost with all of this (-:

PS: I moved the couch a few days ago and put the speakers in a symmetrical arrangement to see if it would help, but nope, same problem. Here is a photo of it:

View attachment 422383
To me it still looks like the speakers are placed much closer to the left wall, but maybe that is just how it looks in the photo?
 
To me it still looks like the speakers are placed much closer to the left wall, but maybe that is just how it looks in the photo?
yes it was on the pic, i moved them more in the center after that. Also tried different toe in/out
 
thats a good idea! Thanks

Should i stick to a "curve" or could i only go down with like 1khz - 1.5 khz, use it like a EQ ?
Perhaps share a screenshot of the Dirac measurements for some guidance on where the harshness might be coming from.

yes it was on the pic, i moved them more in the center after that. Also tried different toe in/out
Avoid moving the speakers after completing your Dirac measurements! It will throw everything off, and you won’t know where you stand or how to adjust the EQ based on the measurements you’ve taken.
 
If you’ve moved the speakers after completing your Dirac measurements, you can no longer use them. You’ll need to redo the measurement procedure from the beginning.
 
heres the REW file, should be better as a picture, right? Dirac first try


Yes, i know. I moved them before dirac or EQ
Actually, I'm more curious about how the merged response curve appears within the Dirac Live program. REW measurements don’t provide the complete picture of how Dirac interprets or processes the multiple measurements.
 
within the Dirac Live program.
not sure which one you'd like to see so i made multiple snapshots. One measure is with 0deg UMIK1 and the other with 90deg UMIK1
 

Attachments

  • Bildschirmfoto 2025-01-19 um 19.17.51.png
    Bildschirmfoto 2025-01-19 um 19.17.51.png
    318 KB · Views: 35
  • Bildschirmfoto 2025-01-19 um 19.17.47.png
    Bildschirmfoto 2025-01-19 um 19.17.47.png
    309.2 KB · Views: 37
  • Bildschirmfoto 2025-01-19 um 19.17.44.png
    Bildschirmfoto 2025-01-19 um 19.17.44.png
    294.3 KB · Views: 35
  • Bildschirmfoto 2025-01-19 um 19.17.02.png
    Bildschirmfoto 2025-01-19 um 19.17.02.png
    307.4 KB · Views: 34
  • Bildschirmfoto 2025-01-19 um 19.16.56.png
    Bildschirmfoto 2025-01-19 um 19.16.56.png
    306.6 KB · Views: 33
  • Bildschirmfoto 2025-01-19 um 19.16.47.png
    Bildschirmfoto 2025-01-19 um 19.16.47.png
    299 KB · Views: 33
Back
Top Bottom